Features

Projects: set up and manage a website in Serpel

A project in Serpel is one website that you analyse. You create it from a domain and a market, and it then holds the crawls, tracked keywords, competitors, reports and settings of that site. Creating a project is free, and nothing is crawled or checked until you start it.

Serpel illustration: an isometric platform with a website block, a blue map pin, a gear and a small orange flag

What is a project in Serpel?

A project is the container for everything about one domain: crawls and the audit score, tracked keywords, competitors, AI prompts, keyword lists, reports, tasks and the connections to Search Console or web analytics. Every project has its own market (a country and a language), its own settings and its own members.

Projects belong to a workspace, and your plan decides how many projects the workspace can hold. Work that runs in a project is charged to the workspace that owns it. Credits and billing explains plans, credits and prices.

project without a subscription
1
projects on Starter
5
projects on Pro
20

How do you create a project in Serpel?

You create a project from a domain. Serpel normalises the domain, creates the project with you as its owner and opens its overview. Account connections to Google, Bing, PostHog or Plausible that you already made under Settings → Integrations are matched to the new project in the background.

From domain to first data
  1. Add the domain

    Enter the domain, an optional name and the market. Nothing is crawled or checked yet.

    Free
  2. Crawl the site

    Start the first crawl from the overview. A crawl costs 1 credit per started 20 pages.

    Credits
  3. Track keywords

    Add the keywords you want to rank for. Adding is free, and the checks cost credits.

    Slots
  4. Connect search data

    Connect Search Console to see real clicks and queries instead of estimates.

    Free
  5. Schedule monitoring

    Choose how often Serpel checks rankings and writes a report.

    Optional
  1. Open the new project dialog.

    Open the project menu at the top of the sidebar and choose New project. On the All projects page you can also use the Project button, or Create first project if you have no project yet.

  2. Enter the domain.

    Type the domain into Domain, for example example.com, or paste a full address. A path in the address becomes the start URL of the crawl.

  3. Name it if you like.

    Name (optional) is only a label. Without it, Serpel uses the domain as the name.

  4. Choose the market.

    Pick the Country and Language that the site targets. The dialog starts with Germany and German.

  5. Create the project.

    Click Create project. Serpel opens the overview of the new project, where the Getting started checklist waits for you.

Which domains does Serpel accept?

Serpel accepts public domain names. It strips www. from the domain, but the start URL for crawls keeps the host, path and query that you typed. Rank checks match your domain and its subdomains.

You enterSerpel storesWhy
Example.comDomain example.com, start URL https://example.com/Names are lower-cased, and an address without a scheme becomes https.
https://www.example.com/shopDomain example.com, start URL https://www.example.com/shopThe www. prefix leaves the domain, and the path becomes the start URL.
shop.example.comDomain shop.example.comA subdomain is treated as its own domain.
example.com:8443RejectedOnly the standard ports 80 and 443 are supported.
192.168.0.10, localhost, names ending in .local or .internalRejectedOnly public domain names can be analysed, not IP addresses or internal hosts.

You cannot create a second project for a domain that you already have a project for. Serpel then names the existing project, and the CLI returns its ID. The domain cannot be changed after creation, so create a new project for a different domain.

Which country and language should you choose?

Serpel supports 43 countries and 22 languages. The market of the project is the default for new tracked keywords and for keyword research. Each tracked keyword keeps the market it was added with, so changing the project market later does not touch existing keywords. See rank tracking for how markets, locations and devices combine.

How many projects can you create?

The limit counts the active projects in your workspace. When you reach it, the dialog shows a usage meter, Create project is disabled and the API answers with HTTP 403 and plan_limit. Deleting a project frees its slot, and a larger plan raises the limit under Settings → Billing. The other per-project limits follow the same plan.

PlanProjectsTracked keywords per projectAI prompts per projectPages per crawlDaily monitoring
No subscription15010250No
Starter5200251,000No
Pro20500501,000Yes

What does the project overview show?

Open a project from All projects or from the project menu in the sidebar to reach its overview. The page fills as you add data, and each section links to the page that explains it:

  • Getting started is the onboarding checklist. It stays until all five steps are done or you click Hide.
  • The monitoring strip shows Monitoring active or Monitoring off, the Frequency, the Last run, the Next run and the Email report status. Latest report opens the newest monitoring report, and Check now starts a monitoring run immediately.
  • Since the last run counts the keywords that moved up or dropped, new page-1 rankings, prompts where ChatGPT cites you and new issues on the website.
  • The opportunities widget lists the best opportunities for the site.
  • Trends charts Visibility, Clicks from Google, Visitors and Audit score for Last 7 days, Last 28 days (the default) or Last 3 months. Charts for Search Console and web analytics appear once those are connected.
  • Cards for clicks, competitors, visitors and positions summarise the other pages, and To do on your website lists the most important issues of the latest crawl.
A project overview with the monitoring strip, the changes since the last run and the trend chart, shown with demo data.

What is on the Getting started checklist?

StepWhat it asks forCounts as done when
Project createdNothing. It links to the project settings.Always, from the start.
Connect Search ConsoleA Google Search Console connection, see search data.A Google connection with a chosen property exists for the project.
Complete your first crawlStart a crawl. The card shows the page limit and the most it can cost, starting at 100 pages by default.A crawl has finished. See site audit.
Check your starting balanceOpen Billing to see your credits.Your workspace has any credit booking, such as the welcome credits.
Start your first keyword researchOpen keyword research.You have started a research in this project.

The checklist updates itself from your real data. Hide stores your choice in a cookie of your browser, so it stays hidden only on that browser.

Which settings does a project have?

Open the project menu in the sidebar and choose Settings. The sidebar then lists the project sections General, Ranking check, Monitoring and Members. Editors and owners can change these settings, except for the domain. After you change a value, a bar with Cancel and Save appears, and nothing is stored until you click Save.

General: name, market and crawl

SettingWhat it does
NameThe label shown in the sidebar and on the project cards.
DomainRead-only. The domain cannot change after creation.
MarketCountry and language, the defaults for new keywords and research.
Pages per crawlThe most pages one crawl may check, from 1 to 1,000 and never more than your plan allows. New projects start with 100.
JavaScript renderingAutomatic for empty pages renders only pages that arrive almost empty, Always render renders every page and Never render reads only the served HTML. See site audit.

Ranking check: how rankings are read

The sidebar calls this section Ranking check, while the heading on the page says “Rank check”. Both are the same settings.

SettingWhat it does
DepthHow many organic results each check reads: top 10, top 20, top 30, top 50, top 100. The deepest level costs more per keyword. New projects start at top 50. See rank tracking.
LocationThe default place for new keywords: all of the country, or a city or postal code. Existing keywords keep their own location.
DeviceThe default device for new keywords: Desktop or Mobile. New projects start on desktop.
Also check Google AI OverviewReads on every check whether Google shows an AI Overview and whether it cites you. It is on by default.
Add search volume automaticallyAdds search volume and CPC to new keywords before a check, and refreshes them every 30 days. It is on by default.

Monitoring: schedules, reports and webhooks

Frequency is Manual only, Daily, Every 3 days or Weekly, and Time is the hour a scheduled run starts. The hour is in Europe/Berlin time, not in your local time. Daily needs the Pro plan. A forecast beneath the fields shows the credits per month, and Weekly crawl adds a crawl every Monday at the same hour. The section also holds the Email report recipients and the Slack and Discord report rows. Monitoring and reports covers all of them.

Who can see and change a project?

Roles belong to a project, not to the workspace. Whoever creates a project becomes its owner. Everyone else is an editor or a viewer.

What each project role can do
Reads

Viewer

  • Sees all data of the project.
  • Cannot change settings, keywords or competitors.
  • Sees a Read only banner on the settings page.
Reads and changes

Editor

  • Starts crawls and checks, maintains keywords and competitors.
  • Changes the project settings.
  • Cannot manage members or delete the project.
Full control

Owner

  • Everything an editor can do.
  • Removes members and sees the Invite member button.
  • Deletes the project.

How do you add members during the beta?

Owners see an Invite member button under Members, with an email address and the roles Viewer and Editor. During the beta, the request only succeeds when a Serpel administrator sends it, so ask the Serpel team to add the person instead. The person needs an existing, active Serpel account, because there are no email invitations. Role changes follow the same rule, and you cannot hand the owner role to someone else.

Owners can remove a member at any time with the trash icon next to the name. The member loses access at once, and the connections that member created for the project are removed too. Everyone can read the list of members, in the dashboard and with serpel projects members list.

How do you delete a project?

  1. Open the project settings.

    Choose Settings in the project menu and scroll to Delete project. Only owners see this section.

  2. Click Delete project.

    A dialog asks you to type the domain of the project as a confirmation.

  3. Confirm.

    Click Delete permanently. Serpel returns you to All projects.

The project disappears at once. Queued jobs are cancelled, running jobs are asked to stop, the schedule is switched off and the project slot becomes free. Serpel removes the stored data for good about 30 days after the deletion. The app has no restore button, so export what you need first with the history and exports tools.

How do you manage projects from the CLI, API and MCP?

Every project action in the dashboard has a CLI command and a REST API endpoint. The commands below create, list, change and delete projects. Changing and deleting follow the same roles as the dashboard.

Project commands
serpel projects create --domain example.com --country US --language en
serpel projects list
serpel projects show <id>
serpel projects update <id> --rank-depth 50 --schedule weekly --schedule-hour 6
serpel projects members list --project <id>
serpel projects delete <id> --yes

The API needs a token with the projects:write scope to create or change a project and projects:read to read one. A token that is limited to certain projects cannot create new ones.

Create a project with the API
curl -X POST https://serpel.app/api/v1/projects \
  -H "Authorization: Bearer $SERPEL_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"domain": "example.com", "name": "Example", "country": "US", "language": "en"}'

The MCP server can read projects but not create or change them. list_projects returns your projects with ID, domain, market and role, and get_project_overview summarises one of them. The Claude Code guide lists all tools.

What if a project does not behave as expected?

What you seeWhat to do
Create project is disabled, or the API answers plan_limitYour workspace has reached its project limit. Delete a project you no longer need or upgrade under Settings → Billing.
“A project for example.com already exists”You already have an active project for that domain. Open the existing project, or delete it before you create another.
“Enter a domain name instead of an IP address”Use the public domain name. IP addresses, localhost and internal hosts cannot be analysed.
The settings are greyed out and a Read only banner showsYou are a viewer in this project. Ask an owner to change your role.
Daily does not saveDaily checks need the Pro plan. Choose Every 3 days or Weekly.
Invite member fails with “project members can only be added by a platform admin”That is the beta rule. Ask the Serpel team to add the person.
The new project is emptyThat is expected. Start a crawl, add keywords or run a research, as in the checklist.

Where to go next

Updated 5 Nov 2026

See also